论文部分内容阅读
采用正交试验研究萃取剂类型、萃取剂体积、分散剂类型、分散剂体积、溶液pH值、离子强度和萃取时间对水样中痕量十溴联苯醚(decaBDE)分散液液微萃取(DLLME)回收率的影响。结果表明,离子强度对萃取回收率(ER)的影响非常显著,而分散剂体积与萃取剂体积交互作用的影响不显著。通过极值法确定的decaBDE分散液液微萃取条件下的萃取回收率为88.24%。以正交试验数据为训练样本,以分散剂体积、萃取剂类型、pH值、离子强度及萃取时间为输入,萃取回收率为输出,建立了影响decaBDE分散液液微萃取的BP神经网络模型。模型检验样本预测输出值和试验值的决定系数为0.8734,表明模型可以预测水样中decaBDE分散液液微萃取的回收率。采用遗传算法工具箱对建立的BP神经网络模型进行优化求解,得到的优化分散液液微萃取条件下的decaBu0003DE萃取回收率平均值为99.94%,比通过极值法确定的萃取回收率提高10%以上。
Orthogonal experiments were carried out to study the effects of extractant type, extractant volume, dispersant type, dispersant volume, solution pH, ionic strength and extraction time on liquid microextraction of decabromodiphenyl ether (decaBDE) in water DLLME) recovery rate. The results showed that the effect of ionic strength on extraction recovery (ER) was significant, while the interaction between dispersant volume and extractant volume was insignificant. The extraction recovery of decaBDE by liquid-liquid microextraction determined by the extremum method was 88.24%. Taking the orthogonal test data as the training sample, taking the volume of dispersant, the type of extractant, the pH value, the ionic strength and the extraction time as the input, the extraction recovery was taken as the output. A BP neural network model was established to influence the decaBDE dispersion liquid microextraction. The coefficient of determination of the predicted output value and the experimental value of the model test sample is 0.8734, indicating that the model can predict the recovery rate of decaBDE dispersion liquid micro-extraction in water sample. The genetic algorithm toolbox was used to optimize the BP neural network model. The average extraction recovery of decaB u0003DE was 99.94%, which was higher than the extraction recovery determined by the extremum method 10% or more.